Hugging Face's logo

stefan-it
/
hmbench-letemps-fr-hmbert_tiny-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
4b08c16
•
1 Parent(s): fca7274

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. runs/events.out.tfevents.1697673483.46dc0c540dd0.3802.4 +3 -0
  5. test.tsv +0 -0
  6. training.log +247 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23a210328680a6bfe55b154dbad6bd53b2eba2c7dbc5dd2e1f18bf7f1ba4855b
3
+ size 19045922
dev.tsv ADDED
The diff for this file is too large to render. See raw diff
 
loss.tsv ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ EPOCH TIMESTAMP LEARNING_RATE TRAIN_LOSS DEV_LOSS DEV_PRECISION DEV_RECALL DEV_F1 DEV_ACCURACY
2
+ 1 23:59:01 0.0000 0.5748 0.1842 0.3172 0.0675 0.1113 0.0602
3
+ 2 00:00:02 0.0000 0.1979 0.1775 0.2514 0.4039 0.3099 0.1910
4
+ 3 00:01:02 0.0000 0.1717 0.1713 0.3069 0.4519 0.3656 0.2340
5
+ 4 00:02:03 0.0000 0.1544 0.1667 0.3404 0.4954 0.4035 0.2648
6
+ 5 00:03:04 0.0000 0.1417 0.1812 0.3510 0.5526 0.4293 0.2880
7
+ 6 00:04:05 0.0000 0.1317 0.1711 0.3934 0.5446 0.4568 0.3101
8
+ 7 00:05:06 0.0000 0.1269 0.1817 0.4008 0.5458 0.4622 0.3140
9
+ 8 00:06:06 0.0000 0.1209 0.1852 0.4010 0.5469 0.4627 0.3130
10
+ 9 00:07:07 0.0000 0.1190 0.1872 0.4028 0.5641 0.4700 0.3203
11
+ 10 00:08:07 0.0000 0.1167 0.1891 0.4021 0.5664 0.4703 0.3196
runs/events.out.tfevents.1697673483.46dc0c540dd0.3802.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb6a3ecd5500b1eafa1ec8717975c503651bc775fa2eb282e34df1a0744975a5
3
+ size 2030580
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,247 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-18 23:58:03,176 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-18 23:58:03,177 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(32001, 128)
7
+ (position_embeddings): Embedding(512, 128)
8
+ (token_type_embeddings): Embedding(2, 128)
9
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
10
+ (dropout): Dropout(p=0.1, inplace=False)
11
+ )
12
+ (encoder): BertEncoder(
13
+ (layer): ModuleList(
14
+ (0-1): 2 x BertLayer(
15
+ (attention): BertAttention(
16
+ (self): BertSelfAttention(
17
+ (query): Linear(in_features=128, out_features=128, bias=True)
18
+ (key): Linear(in_features=128, out_features=128, bias=True)
19
+ (value): Linear(in_features=128, out_features=128, bias=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (output): BertSelfOutput(
23
+ (dense): Linear(in_features=128, out_features=128, bias=True)
24
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
25
+ (dropout): Dropout(p=0.1, inplace=False)
26
+ )
27
+ )
28
+ (intermediate): BertIntermediate(
29
+ (dense): Linear(in_features=128, out_features=512, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): BertOutput(
33
+ (dense): Linear(in_features=512, out_features=128, bias=True)
34
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
35
+ (dropout): Dropout(p=0.1, inplace=False)
36
+ )
37
+ )
38
+ )
39
+ )
40
+ (pooler): BertPooler(
41
+ (dense): Linear(in_features=128, out_features=128, bias=True)
42
+ (activation): Tanh()
43
+ )
44
+ )
45
+ )
46
+ (locked_dropout): LockedDropout(p=0.5)
47
+ (linear): Linear(in_features=128, out_features=13, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-18 23:58:03,177 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-18 23:58:03,177 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences
52
+ - NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator
53
+ 2023-10-18 23:58:03,177 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-18 23:58:03,177 Train: 14465 sentences
55
+ 2023-10-18 23:58:03,177 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-18 23:58:03,177 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-18 23:58:03,177 Training Params:
58
+ 2023-10-18 23:58:03,177 - learning_rate: "3e-05"
59
+ 2023-10-18 23:58:03,177 - mini_batch_size: "4"
60
+ 2023-10-18 23:58:03,177 - max_epochs: "10"
61
+ 2023-10-18 23:58:03,177 - shuffle: "True"
62
+ 2023-10-18 23:58:03,177 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-18 23:58:03,177 Plugins:
64
+ 2023-10-18 23:58:03,177 - TensorboardLogger
65
+ 2023-10-18 23:58:03,177 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-18 23:58:03,177 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-18 23:58:03,177 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-18 23:58:03,177 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-18 23:58:03,177 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-18 23:58:03,177 Computation:
71
+ 2023-10-18 23:58:03,177 - compute on device: cuda:0
72
+ 2023-10-18 23:58:03,177 - embedding storage: none
73
+ 2023-10-18 23:58:03,177 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-18 23:58:03,177 Model training base path: "hmbench-letemps/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
75
+ 2023-10-18 23:58:03,177 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-18 23:58:03,178 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-18 23:58:03,178 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-18 23:58:08,803 epoch 1 - iter 361/3617 - loss 2.36997878 - time (sec): 5.63 - samples/sec: 6809.84 - lr: 0.000003 - momentum: 0.000000
79
+ 2023-10-18 23:58:14,467 epoch 1 - iter 722/3617 - loss 1.77258874 - time (sec): 11.29 - samples/sec: 6600.74 - lr: 0.000006 - momentum: 0.000000
80
+ 2023-10-18 23:58:20,123 epoch 1 - iter 1083/3617 - loss 1.31516970 - time (sec): 16.94 - samples/sec: 6619.34 - lr: 0.000009 - momentum: 0.000000
81
+ 2023-10-18 23:58:25,656 epoch 1 - iter 1444/3617 - loss 1.06703131 - time (sec): 22.48 - samples/sec: 6702.33 - lr: 0.000012 - momentum: 0.000000
82
+ 2023-10-18 23:58:31,174 epoch 1 - iter 1805/3617 - loss 0.91344974 - time (sec): 28.00 - samples/sec: 6727.80 - lr: 0.000015 - momentum: 0.000000
83
+ 2023-10-18 23:58:36,755 epoch 1 - iter 2166/3617 - loss 0.81603607 - time (sec): 33.58 - samples/sec: 6675.28 - lr: 0.000018 - momentum: 0.000000
84
+ 2023-10-18 23:58:42,404 epoch 1 - iter 2527/3617 - loss 0.73338690 - time (sec): 39.23 - samples/sec: 6673.64 - lr: 0.000021 - momentum: 0.000000
85
+ 2023-10-18 23:58:48,315 epoch 1 - iter 2888/3617 - loss 0.66902349 - time (sec): 45.14 - samples/sec: 6667.42 - lr: 0.000024 - momentum: 0.000000
86
+ 2023-10-18 23:58:53,645 epoch 1 - iter 3249/3617 - loss 0.61751533 - time (sec): 50.47 - samples/sec: 6734.11 - lr: 0.000027 - momentum: 0.000000
87
+ 2023-10-18 23:58:58,925 epoch 1 - iter 3610/3617 - loss 0.57566262 - time (sec): 55.75 - samples/sec: 6803.03 - lr: 0.000030 - momentum: 0.000000
88
+ 2023-10-18 23:58:59,030 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-18 23:58:59,030 EPOCH 1 done: loss 0.5748 - lr: 0.000030
90
+ 2023-10-18 23:59:01,374 DEV : loss 0.18418805301189423 - f1-score (micro avg) 0.1113
91
+ 2023-10-18 23:59:01,406 saving best model
92
+ 2023-10-18 23:59:01,439 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-18 23:59:07,237 epoch 2 - iter 361/3617 - loss 0.21498018 - time (sec): 5.80 - samples/sec: 6656.89 - lr: 0.000030 - momentum: 0.000000
94
+ 2023-10-18 23:59:12,737 epoch 2 - iter 722/3617 - loss 0.21259486 - time (sec): 11.30 - samples/sec: 6753.82 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-18 23:59:18,069 epoch 2 - iter 1083/3617 - loss 0.21274781 - time (sec): 16.63 - samples/sec: 6784.57 - lr: 0.000029 - momentum: 0.000000
96
+ 2023-10-18 23:59:23,748 epoch 2 - iter 1444/3617 - loss 0.21183692 - time (sec): 22.31 - samples/sec: 6787.56 - lr: 0.000029 - momentum: 0.000000
97
+ 2023-10-18 23:59:29,474 epoch 2 - iter 1805/3617 - loss 0.20972085 - time (sec): 28.03 - samples/sec: 6761.65 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-18 23:59:35,061 epoch 2 - iter 2166/3617 - loss 0.20622936 - time (sec): 33.62 - samples/sec: 6745.79 - lr: 0.000028 - momentum: 0.000000
99
+ 2023-10-18 23:59:40,816 epoch 2 - iter 2527/3617 - loss 0.20279605 - time (sec): 39.38 - samples/sec: 6723.03 - lr: 0.000028 - momentum: 0.000000
100
+ 2023-10-18 23:59:46,581 epoch 2 - iter 2888/3617 - loss 0.20277512 - time (sec): 45.14 - samples/sec: 6715.36 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-18 23:59:52,401 epoch 2 - iter 3249/3617 - loss 0.19959401 - time (sec): 50.96 - samples/sec: 6718.71 - lr: 0.000027 - momentum: 0.000000
102
+ 2023-10-18 23:59:58,072 epoch 2 - iter 3610/3617 - loss 0.19787867 - time (sec): 56.63 - samples/sec: 6698.97 - lr: 0.000027 - momentum: 0.000000
103
+ 2023-10-18 23:59:58,174 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-18 23:59:58,174 EPOCH 2 done: loss 0.1979 - lr: 0.000027
105
+ 2023-10-19 00:00:02,011 DEV : loss 0.17745766043663025 - f1-score (micro avg) 0.3099
106
+ 2023-10-19 00:00:02,038 saving best model
107
+ 2023-10-19 00:00:02,071 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-19 00:00:07,839 epoch 3 - iter 361/3617 - loss 0.17426994 - time (sec): 5.77 - samples/sec: 6426.35 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-19 00:00:13,678 epoch 3 - iter 722/3617 - loss 0.18366482 - time (sec): 11.61 - samples/sec: 6425.53 - lr: 0.000026 - momentum: 0.000000
110
+ 2023-10-19 00:00:19,477 epoch 3 - iter 1083/3617 - loss 0.18265066 - time (sec): 17.41 - samples/sec: 6527.22 - lr: 0.000026 - momentum: 0.000000
111
+ 2023-10-19 00:00:25,181 epoch 3 - iter 1444/3617 - loss 0.17944185 - time (sec): 23.11 - samples/sec: 6560.81 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-19 00:00:30,946 epoch 3 - iter 1805/3617 - loss 0.17792967 - time (sec): 28.87 - samples/sec: 6492.08 - lr: 0.000025 - momentum: 0.000000
113
+ 2023-10-19 00:00:36,733 epoch 3 - iter 2166/3617 - loss 0.17586984 - time (sec): 34.66 - samples/sec: 6505.40 - lr: 0.000025 - momentum: 0.000000
114
+ 2023-10-19 00:00:42,495 epoch 3 - iter 2527/3617 - loss 0.17550889 - time (sec): 40.42 - samples/sec: 6522.07 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-19 00:00:47,896 epoch 3 - iter 2888/3617 - loss 0.17374263 - time (sec): 45.82 - samples/sec: 6581.15 - lr: 0.000024 - momentum: 0.000000
116
+ 2023-10-19 00:00:53,585 epoch 3 - iter 3249/3617 - loss 0.17360638 - time (sec): 51.51 - samples/sec: 6613.70 - lr: 0.000024 - momentum: 0.000000
117
+ 2023-10-19 00:00:59,356 epoch 3 - iter 3610/3617 - loss 0.17167513 - time (sec): 57.28 - samples/sec: 6619.96 - lr: 0.000023 - momentum: 0.000000
118
+ 2023-10-19 00:00:59,468 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-19 00:00:59,468 EPOCH 3 done: loss 0.1717 - lr: 0.000023
120
+ 2023-10-19 00:01:02,699 DEV : loss 0.1712975651025772 - f1-score (micro avg) 0.3656
121
+ 2023-10-19 00:01:02,726 saving best model
122
+ 2023-10-19 00:01:02,762 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-19 00:01:08,557 epoch 4 - iter 361/3617 - loss 0.17147098 - time (sec): 5.79 - samples/sec: 6526.12 - lr: 0.000023 - momentum: 0.000000
124
+ 2023-10-19 00:01:14,379 epoch 4 - iter 722/3617 - loss 0.16727696 - time (sec): 11.62 - samples/sec: 6680.36 - lr: 0.000023 - momentum: 0.000000
125
+ 2023-10-19 00:01:20,265 epoch 4 - iter 1083/3617 - loss 0.15984749 - time (sec): 17.50 - samples/sec: 6583.25 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-19 00:01:25,975 epoch 4 - iter 1444/3617 - loss 0.15449183 - time (sec): 23.21 - samples/sec: 6553.37 - lr: 0.000022 - momentum: 0.000000
127
+ 2023-10-19 00:01:31,736 epoch 4 - iter 1805/3617 - loss 0.15441485 - time (sec): 28.97 - samples/sec: 6545.80 - lr: 0.000022 - momentum: 0.000000
128
+ 2023-10-19 00:01:37,433 epoch 4 - iter 2166/3617 - loss 0.15394200 - time (sec): 34.67 - samples/sec: 6594.35 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-19 00:01:42,927 epoch 4 - iter 2527/3617 - loss 0.15538219 - time (sec): 40.16 - samples/sec: 6635.54 - lr: 0.000021 - momentum: 0.000000
130
+ 2023-10-19 00:01:48,596 epoch 4 - iter 2888/3617 - loss 0.15441080 - time (sec): 45.83 - samples/sec: 6624.20 - lr: 0.000021 - momentum: 0.000000
131
+ 2023-10-19 00:01:54,263 epoch 4 - iter 3249/3617 - loss 0.15565106 - time (sec): 51.50 - samples/sec: 6621.64 - lr: 0.000020 - momentum: 0.000000
132
+ 2023-10-19 00:02:00,045 epoch 4 - iter 3610/3617 - loss 0.15428029 - time (sec): 57.28 - samples/sec: 6620.88 - lr: 0.000020 - momentum: 0.000000
133
+ 2023-10-19 00:02:00,158 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-19 00:02:00,158 EPOCH 4 done: loss 0.1544 - lr: 0.000020
135
+ 2023-10-19 00:02:03,402 DEV : loss 0.16668201982975006 - f1-score (micro avg) 0.4035
136
+ 2023-10-19 00:02:03,430 saving best model
137
+ 2023-10-19 00:02:03,466 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-19 00:02:09,207 epoch 5 - iter 361/3617 - loss 0.12821978 - time (sec): 5.74 - samples/sec: 6756.08 - lr: 0.000020 - momentum: 0.000000
139
+ 2023-10-19 00:02:14,914 epoch 5 - iter 722/3617 - loss 0.13135493 - time (sec): 11.45 - samples/sec: 6814.87 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-19 00:02:20,535 epoch 5 - iter 1083/3617 - loss 0.12991170 - time (sec): 17.07 - samples/sec: 6702.45 - lr: 0.000019 - momentum: 0.000000
141
+ 2023-10-19 00:02:26,234 epoch 5 - iter 1444/3617 - loss 0.13451604 - time (sec): 22.77 - samples/sec: 6671.61 - lr: 0.000019 - momentum: 0.000000
142
+ 2023-10-19 00:02:31,990 epoch 5 - iter 1805/3617 - loss 0.13846108 - time (sec): 28.52 - samples/sec: 6607.20 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-19 00:02:37,703 epoch 5 - iter 2166/3617 - loss 0.13989470 - time (sec): 34.24 - samples/sec: 6625.27 - lr: 0.000018 - momentum: 0.000000
144
+ 2023-10-19 00:02:43,353 epoch 5 - iter 2527/3617 - loss 0.14171266 - time (sec): 39.89 - samples/sec: 6608.77 - lr: 0.000018 - momentum: 0.000000
145
+ 2023-10-19 00:02:48,838 epoch 5 - iter 2888/3617 - loss 0.14217721 - time (sec): 45.37 - samples/sec: 6650.03 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-19 00:02:54,617 epoch 5 - iter 3249/3617 - loss 0.14146502 - time (sec): 51.15 - samples/sec: 6645.97 - lr: 0.000017 - momentum: 0.000000
147
+ 2023-10-19 00:03:00,552 epoch 5 - iter 3610/3617 - loss 0.14181052 - time (sec): 57.08 - samples/sec: 6643.68 - lr: 0.000017 - momentum: 0.000000
148
+ 2023-10-19 00:03:00,655 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-19 00:03:00,655 EPOCH 5 done: loss 0.1417 - lr: 0.000017
150
+ 2023-10-19 00:03:04,530 DEV : loss 0.18115580081939697 - f1-score (micro avg) 0.4293
151
+ 2023-10-19 00:03:04,560 saving best model
152
+ 2023-10-19 00:03:04,596 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-19 00:03:10,102 epoch 6 - iter 361/3617 - loss 0.14639806 - time (sec): 5.51 - samples/sec: 6548.38 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-19 00:03:15,855 epoch 6 - iter 722/3617 - loss 0.13502357 - time (sec): 11.26 - samples/sec: 6600.08 - lr: 0.000016 - momentum: 0.000000
155
+ 2023-10-19 00:03:21,611 epoch 6 - iter 1083/3617 - loss 0.13028085 - time (sec): 17.01 - samples/sec: 6543.90 - lr: 0.000016 - momentum: 0.000000
156
+ 2023-10-19 00:03:27,331 epoch 6 - iter 1444/3617 - loss 0.13278529 - time (sec): 22.74 - samples/sec: 6549.58 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-19 00:03:33,378 epoch 6 - iter 1805/3617 - loss 0.13346841 - time (sec): 28.78 - samples/sec: 6538.15 - lr: 0.000015 - momentum: 0.000000
158
+ 2023-10-19 00:03:39,165 epoch 6 - iter 2166/3617 - loss 0.13376779 - time (sec): 34.57 - samples/sec: 6533.47 - lr: 0.000015 - momentum: 0.000000
159
+ 2023-10-19 00:03:44,876 epoch 6 - iter 2527/3617 - loss 0.13205667 - time (sec): 40.28 - samples/sec: 6534.50 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-19 00:03:50,571 epoch 6 - iter 2888/3617 - loss 0.13254041 - time (sec): 45.97 - samples/sec: 6528.00 - lr: 0.000014 - momentum: 0.000000
161
+ 2023-10-19 00:03:56,310 epoch 6 - iter 3249/3617 - loss 0.13140323 - time (sec): 51.71 - samples/sec: 6560.47 - lr: 0.000014 - momentum: 0.000000
162
+ 2023-10-19 00:04:02,061 epoch 6 - iter 3610/3617 - loss 0.13177254 - time (sec): 57.46 - samples/sec: 6601.78 - lr: 0.000013 - momentum: 0.000000
163
+ 2023-10-19 00:04:02,157 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-19 00:04:02,157 EPOCH 6 done: loss 0.1317 - lr: 0.000013
165
+ 2023-10-19 00:04:05,350 DEV : loss 0.17110641300678253 - f1-score (micro avg) 0.4568
166
+ 2023-10-19 00:04:05,378 saving best model
167
+ 2023-10-19 00:04:05,412 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-19 00:04:11,107 epoch 7 - iter 361/3617 - loss 0.12258612 - time (sec): 5.69 - samples/sec: 6561.55 - lr: 0.000013 - momentum: 0.000000
169
+ 2023-10-19 00:04:16,503 epoch 7 - iter 722/3617 - loss 0.12769325 - time (sec): 11.09 - samples/sec: 6782.46 - lr: 0.000013 - momentum: 0.000000
170
+ 2023-10-19 00:04:22,269 epoch 7 - iter 1083/3617 - loss 0.12762667 - time (sec): 16.86 - samples/sec: 6690.09 - lr: 0.000012 - momentum: 0.000000
171
+ 2023-10-19 00:04:27,932 epoch 7 - iter 1444/3617 - loss 0.13095263 - time (sec): 22.52 - samples/sec: 6642.13 - lr: 0.000012 - momentum: 0.000000
172
+ 2023-10-19 00:04:33,641 epoch 7 - iter 1805/3617 - loss 0.13129488 - time (sec): 28.23 - samples/sec: 6607.53 - lr: 0.000012 - momentum: 0.000000
173
+ 2023-10-19 00:04:39,080 epoch 7 - iter 2166/3617 - loss 0.12959570 - time (sec): 33.67 - samples/sec: 6698.16 - lr: 0.000011 - momentum: 0.000000
174
+ 2023-10-19 00:04:44,807 epoch 7 - iter 2527/3617 - loss 0.12880100 - time (sec): 39.39 - samples/sec: 6687.27 - lr: 0.000011 - momentum: 0.000000
175
+ 2023-10-19 00:04:50,566 epoch 7 - iter 2888/3617 - loss 0.12959427 - time (sec): 45.15 - samples/sec: 6685.86 - lr: 0.000011 - momentum: 0.000000
176
+ 2023-10-19 00:04:56,329 epoch 7 - iter 3249/3617 - loss 0.12780021 - time (sec): 50.92 - samples/sec: 6695.46 - lr: 0.000010 - momentum: 0.000000
177
+ 2023-10-19 00:05:02,113 epoch 7 - iter 3610/3617 - loss 0.12697124 - time (sec): 56.70 - samples/sec: 6685.95 - lr: 0.000010 - momentum: 0.000000
178
+ 2023-10-19 00:05:02,216 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-19 00:05:02,217 EPOCH 7 done: loss 0.1269 - lr: 0.000010
180
+ 2023-10-19 00:05:06,100 DEV : loss 0.1817493438720703 - f1-score (micro avg) 0.4622
181
+ 2023-10-19 00:05:06,128 saving best model
182
+ 2023-10-19 00:05:06,166 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-19 00:05:11,879 epoch 8 - iter 361/3617 - loss 0.11567923 - time (sec): 5.71 - samples/sec: 6721.37 - lr: 0.000010 - momentum: 0.000000
184
+ 2023-10-19 00:05:17,542 epoch 8 - iter 722/3617 - loss 0.11726755 - time (sec): 11.38 - samples/sec: 6656.67 - lr: 0.000009 - momentum: 0.000000
185
+ 2023-10-19 00:05:23,292 epoch 8 - iter 1083/3617 - loss 0.11848005 - time (sec): 17.13 - samples/sec: 6656.08 - lr: 0.000009 - momentum: 0.000000
186
+ 2023-10-19 00:05:29,007 epoch 8 - iter 1444/3617 - loss 0.12044334 - time (sec): 22.84 - samples/sec: 6662.48 - lr: 0.000009 - momentum: 0.000000
187
+ 2023-10-19 00:05:34,761 epoch 8 - iter 1805/3617 - loss 0.12046610 - time (sec): 28.59 - samples/sec: 6689.38 - lr: 0.000008 - momentum: 0.000000
188
+ 2023-10-19 00:05:40,271 epoch 8 - iter 2166/3617 - loss 0.12042788 - time (sec): 34.10 - samples/sec: 6663.89 - lr: 0.000008 - momentum: 0.000000
189
+ 2023-10-19 00:05:45,980 epoch 8 - iter 2527/3617 - loss 0.11893839 - time (sec): 39.81 - samples/sec: 6641.06 - lr: 0.000008 - momentum: 0.000000
190
+ 2023-10-19 00:05:51,718 epoch 8 - iter 2888/3617 - loss 0.11857905 - time (sec): 45.55 - samples/sec: 6625.96 - lr: 0.000007 - momentum: 0.000000
191
+ 2023-10-19 00:05:57,262 epoch 8 - iter 3249/3617 - loss 0.12005226 - time (sec): 51.10 - samples/sec: 6673.14 - lr: 0.000007 - momentum: 0.000000
192
+ 2023-10-19 00:06:02,972 epoch 8 - iter 3610/3617 - loss 0.12081155 - time (sec): 56.81 - samples/sec: 6675.12 - lr: 0.000007 - momentum: 0.000000
193
+ 2023-10-19 00:06:03,084 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-19 00:06:03,084 EPOCH 8 done: loss 0.1209 - lr: 0.000007
195
+ 2023-10-19 00:06:06,316 DEV : loss 0.18522560596466064 - f1-score (micro avg) 0.4627
196
+ 2023-10-19 00:06:06,345 saving best model
197
+ 2023-10-19 00:06:06,383 ----------------------------------------------------------------------------------------------------
198
+ 2023-10-19 00:06:12,202 epoch 9 - iter 361/3617 - loss 0.11578345 - time (sec): 5.82 - samples/sec: 6446.83 - lr: 0.000006 - momentum: 0.000000
199
+ 2023-10-19 00:06:17,868 epoch 9 - iter 722/3617 - loss 0.11658425 - time (sec): 11.48 - samples/sec: 6471.78 - lr: 0.000006 - momentum: 0.000000
200
+ 2023-10-19 00:06:23,568 epoch 9 - iter 1083/3617 - loss 0.11552563 - time (sec): 17.18 - samples/sec: 6563.38 - lr: 0.000006 - momentum: 0.000000
201
+ 2023-10-19 00:06:29,261 epoch 9 - iter 1444/3617 - loss 0.11439846 - time (sec): 22.88 - samples/sec: 6626.62 - lr: 0.000005 - momentum: 0.000000
202
+ 2023-10-19 00:06:35,034 epoch 9 - iter 1805/3617 - loss 0.11471886 - time (sec): 28.65 - samples/sec: 6609.74 - lr: 0.000005 - momentum: 0.000000
203
+ 2023-10-19 00:06:40,941 epoch 9 - iter 2166/3617 - loss 0.11616035 - time (sec): 34.56 - samples/sec: 6631.29 - lr: 0.000005 - momentum: 0.000000
204
+ 2023-10-19 00:06:46,712 epoch 9 - iter 2527/3617 - loss 0.11804181 - time (sec): 40.33 - samples/sec: 6595.53 - lr: 0.000004 - momentum: 0.000000
205
+ 2023-10-19 00:06:52,350 epoch 9 - iter 2888/3617 - loss 0.11740974 - time (sec): 45.97 - samples/sec: 6639.79 - lr: 0.000004 - momentum: 0.000000
206
+ 2023-10-19 00:06:57,979 epoch 9 - iter 3249/3617 - loss 0.11893294 - time (sec): 51.60 - samples/sec: 6632.09 - lr: 0.000004 - momentum: 0.000000
207
+ 2023-10-19 00:07:03,694 epoch 9 - iter 3610/3617 - loss 0.11887323 - time (sec): 57.31 - samples/sec: 6621.10 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-19 00:07:03,798 ----------------------------------------------------------------------------------------------------
209
+ 2023-10-19 00:07:03,798 EPOCH 9 done: loss 0.1190 - lr: 0.000003
210
+ 2023-10-19 00:07:07,005 DEV : loss 0.18721944093704224 - f1-score (micro avg) 0.47
211
+ 2023-10-19 00:07:07,036 saving best model
212
+ 2023-10-19 00:07:07,071 ----------------------------------------------------------------------------------------------------
213
+ 2023-10-19 00:07:12,683 epoch 10 - iter 361/3617 - loss 0.11147538 - time (sec): 5.61 - samples/sec: 6402.00 - lr: 0.000003 - momentum: 0.000000
214
+ 2023-10-19 00:07:18,285 epoch 10 - iter 722/3617 - loss 0.11257207 - time (sec): 11.21 - samples/sec: 6642.89 - lr: 0.000003 - momentum: 0.000000
215
+ 2023-10-19 00:07:23,989 epoch 10 - iter 1083/3617 - loss 0.11426618 - time (sec): 16.92 - samples/sec: 6584.35 - lr: 0.000002 - momentum: 0.000000
216
+ 2023-10-19 00:07:29,688 epoch 10 - iter 1444/3617 - loss 0.11495492 - time (sec): 22.62 - samples/sec: 6615.37 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-19 00:07:35,257 epoch 10 - iter 1805/3617 - loss 0.11633593 - time (sec): 28.19 - samples/sec: 6670.70 - lr: 0.000002 - momentum: 0.000000
218
+ 2023-10-19 00:07:41,001 epoch 10 - iter 2166/3617 - loss 0.11656673 - time (sec): 33.93 - samples/sec: 6654.08 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-19 00:07:46,840 epoch 10 - iter 2527/3617 - loss 0.11556910 - time (sec): 39.77 - samples/sec: 6663.47 - lr: 0.000001 - momentum: 0.000000
220
+ 2023-10-19 00:07:52,358 epoch 10 - iter 2888/3617 - loss 0.11448283 - time (sec): 45.29 - samples/sec: 6683.36 - lr: 0.000001 - momentum: 0.000000
221
+ 2023-10-19 00:07:58,097 epoch 10 - iter 3249/3617 - loss 0.11653997 - time (sec): 51.03 - samples/sec: 6710.72 - lr: 0.000000 - momentum: 0.000000
222
+ 2023-10-19 00:08:03,672 epoch 10 - iter 3610/3617 - loss 0.11682170 - time (sec): 56.60 - samples/sec: 6699.07 - lr: 0.000000 - momentum: 0.000000
223
+ 2023-10-19 00:08:03,781 ----------------------------------------------------------------------------------------------------
224
+ 2023-10-19 00:08:03,782 EPOCH 10 done: loss 0.1167 - lr: 0.000000
225
+ 2023-10-19 00:08:07,665 DEV : loss 0.18911151587963104 - f1-score (micro avg) 0.4703
226
+ 2023-10-19 00:08:07,693 saving best model
227
+ 2023-10-19 00:08:07,758 ----------------------------------------------------------------------------------------------------
228
+ 2023-10-19 00:08:07,758 Loading model from best epoch ...
229
+ 2023-10-19 00:08:07,838 SequenceTagger predicts: Dictionary with 13 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org
230
+ 2023-10-19 00:08:11,239
231
+ Results:
232
+ - F-score (micro) 0.5026
233
+ - F-score (macro) 0.3266
234
+ - Accuracy 0.3503
235
+
236
+ By class:
237
+ precision recall f1-score support
238
+
239
+ loc 0.5060 0.7140 0.5923 591
240
+ pers 0.3636 0.4146 0.3874 357
241
+ org 0.0000 0.0000 0.0000 79
242
+
243
+ micro avg 0.4593 0.5550 0.5026 1027
244
+ macro avg 0.2899 0.3762 0.3266 1027
245
+ weighted avg 0.4176 0.5550 0.4755 1027
246
+
247
+ 2023-10-19 00:08:11,239 ----------------------------------------------------------------------------------------------------